Conversation
4df4911 to
62da376
Compare
|
Hi @joews, thanks for taking this on, and sorry for the late follow-up. The goal is to model this very closely to V8's StackFrame interface which states:
I take this to mean that the Regarding the "parent" relationship, in my mind the "outer" script is considered the "parent" and the "inner" script (eval'd code) is the "child". The reasoning loosely is that you get closer and closer to the actual source as you go down the depth of the StackFrame tree branch. I'm willing to change the model a bit if you have a better idea, though. I appreciate you locking down the behavior with tests 👍 Please let me know if you have other questions, and I promise I'll be quicker to respond. |
|
If this is still an issue, can you resolve the conflict and ping me back once done? |
Work in progress fix for #32. I'm opening the PR early to check the approach and ask some questions 😄
So far I have hacked in basic v8 support and tested on Chrome 58. I'll do the other engines and eval tests when i'm happy with the approach.
A few questions about
setEvalOrigin. Given this program:evalin index.js orwillThrowin<anonymous>?isEvalflag?The logic in the PR so far is a simple delta from master which has
willThrow/index.jsas the parent.Description
Expose inner line and column information for stack frames inside
evalcode.Motivation and Context
See #32.
How Has This Been Tested?
So far: one unit test on a stack frame from Chrome 58.
Types of changes
Checklist:
node_modules/.bin/jscs -c .jscsrc error-stack-parser.jspasses without errorsnpm testpasses without errors